The Objective Is the Bottleneck: Latent World Models Encode What Their Planners Cannot Use
作者: Joyjeet Singh
分类: cs.LG, cs.AI
发布日期: 2026-08-13
备注: Follow-up to arXiv:2608.10145. All experiments run on a laptop CPU; no model was trained or fine-tuned. Code, checkpoints and every measurement: github.com/joyjeet-singh/tinylab
💡 一句话要点
提出新的目标函数以解决长时间规划中的瓶颈问题
🎯 匹配领域: 支柱二：RL算法与架构 (RL & Architecture)
关键词: 潜在世界模型 长时间规划 目标函数优化 机器人导航 自动驾驶
📋 核心要点
- 现有的潜在世界模型在长时间规划中表现不佳，通常被认为是预测器的性能限制了规划效果。
- 论文提出通过替换目标函数来改善规划效果，强调目标函数的选择对长时间规划成功率的影响。
- 实验结果显示，替换目标函数后，成功率从26%提升至98%，并且在预算限制下依然表现优异。
📝 摘要（中文）
潜在世界模型的评估通常依赖于其预测能力，但当长时间规划失败时，往往认为预测器的性能下降。本文通过对LeWorldModel在TwoRoom环境中的重现，指出瓶颈在于规划者的目标函数而非预测器。实验表明，替换目标函数后，成功率显著提升，从26%提高至98%。此外，论文还探讨了不同目标函数对规划效果的影响，强调了规划的目标选择对成功率的重要性。
🔬 方法详解
问题定义：本文旨在解决潜在世界模型在长时间规划中的瓶颈问题，现有方法常常将失败归因于预测器的性能不足。
核心思路：论文提出通过优化规划者的目标函数来改善长时间规划的效果，强调目标函数的设计对规划成功率的影响。
技术框架：研究中使用了LeWorldModel的重现，重点分析了目标函数的选择与规划效果之间的关系。主要模块包括预测器、规划器和目标函数设计。
关键创新：最重要的创新在于识别出规划者的目标函数是限制长时间规划成功的关键，而非预测器的能力。通过替换目标函数，显著提升了规划效果。
关键设计：在实验中，使用了交叉熵方法进行规划，优化了目标函数的设计，确保其能够有效降低规划成本。实验还表明，规划的成功与目标函数的选择密切相关。
🖼️ 关键图片
📊 实验亮点
实验结果显示，替换目标函数后，在偏移量为100时，成功率从26%提升至98%，并且在预算限制下，成功率仍然保持在92%。这一结果表明，优化目标函数能够显著改善长时间规划的效果，超越传统方法的限制。
🎯 应用场景
该研究的潜在应用场景包括机器人导航、自动驾驶和智能决策系统等领域。通过优化目标函数，可以提升这些系统在复杂环境中的规划能力，从而提高其自主性和效率。未来，该方法可能推动更智能的规划算法的发展，适用于更广泛的应用场景。
📄 摘要（原文）
Latent world models are judged by how well they predict, so when planning fails at long horizons the natural reading is that the predictor degrades. On a reproduction of LeWorldModel on TwoRoom we show the binding constraint is the planner's objective instead. The predictor is not the limit: its imagined state seventy-five environment steps ahead is still only 0.189 as wrong as assuming the world froze, while the planner never imagines beyond twenty-five. The objective is. Cross-entropy-method planning minimises squared latent distance, which tracks true distance at r = 0.426, saturates by about eighty arena units and decreases beyond a hundred and twenty, so moving away from the goal can lower the cost. The information is present throughout: a ridge probe recovers position from the frozen embedding at R^2 0.9922. The pathology is the method's, not one reimplementation's. It is present in the authors' released weights, and across four checkpoints long-horizon success rank-orders exactly with metric quality and inversely with prediction accuracy. Replacing only the objective, with nothing retrained and no GPU, lifts goals reached at offset 100 from 26.0% to 98.0%, equals the 98.0% at offset 25, and reaches 92.0% under a third of the budget: planning stops depending on the horizon. The best cost is not the most accurate. A head learned from frame separation alone predicts spatial distance worse than a position probe (r = 0.819 against 0.9897) yet plans better, charging 24% more to cross the environment's dividing wall where squared latent distance charges 4% less. It has learned reachability, not proximity.